How to beginner · 3 min read

LiteLLM supported providers list

Q: LiteLLM supported providers list

The LiteLLM Python library supports multiple AI providers including OpenAI, Anthropic, Google Gemini, Meta LLaMA, Mistral, and DeepSeek. This enables developers to switch between providers easily using a unified interface.

Quick answer

The LiteLLM Python library supports multiple AI providers including OpenAI, Anthropic, Google Gemini, Meta LLaMA, Mistral, and DeepSeek. This enables developers to switch between providers easily using a unified interface.

PREREQUISITES

Python 3.8+
API keys for chosen AI providers
pip install litellm

Setup

Install LiteLLM via pip and set environment variables for your AI provider API keys.

Run pip install litellm
Set environment variables like OPENAI_API_KEY, ANTHROPIC_API_KEY, etc.

bash

pip install litellm

Step by step

Use LiteLLM to instantiate clients for supported providers by specifying the provider name and API key. Below is an example for OpenAI and Anthropic.

python

import os
from litellm import LiteLLM

# Initialize OpenAI provider
openai_client = LiteLLM(provider="openai", api_key=os.environ["OPENAI_API_KEY"])
response_openai = openai_client.chat_completion(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello from OpenAI!"}]
)
print("OpenAI response:", response_openai.choices[0].message.content)

# Initialize Anthropic provider
anthropic_client = LiteLLM(provider="anthropic", api_key=os.environ["ANTHROPIC_API_KEY"])
response_anthropic = anthropic_client.chat_completion(
    model="claude-3-5-sonnet-20241022",
    system="You are a helpful assistant.",
    messages=[{"role": "user", "content": "Hello from Anthropic!"}]
)
print("Anthropic response:", response_anthropic.choices[0].message.content)

output

OpenAI response: Hello from OpenAI!
Anthropic response: Hello from Anthropic!

Common variations

LiteLLM supports additional providers such as Google Gemini, Meta LLaMA, Mistral, and DeepSeek. You can switch providers by changing the provider parameter. Async usage and streaming are also supported depending on the provider.

python

import asyncio
from litellm import LiteLLM

async def async_example():
    gemini_client = LiteLLM(provider="google_gemini", api_key=os.environ["GOOGLE_API_KEY"])
    response = await gemini_client.chat_completion_async(
        model="gemini-2.5-pro",
        messages=[{"role": "user", "content": "Hello from Gemini!"}]
    )
    print("Gemini async response:", response.choices[0].message.content)

asyncio.run(async_example())

output

Gemini async response: Hello from Gemini!

Troubleshooting

If you encounter authentication errors, verify your API keys are correctly set in environment variables. For unsupported models or providers, check the LiteLLM documentation for the latest supported list. Network issues may require retry logic or proxy configuration.

✅

Key Takeaways

LiteLLM supports major AI providers including OpenAI, Anthropic, Google Gemini, Meta LLaMA, Mistral, and DeepSeek.
Switch providers easily by changing the provider parameter in LiteLLM client initialization.
Ensure API keys are set in environment variables for seamless authentication.
Async and streaming calls are supported depending on the provider.
Check LiteLLM docs regularly as supported providers and models may update.

Verified 2026-04 · gpt-4o, claude-3-5-sonnet-20241022, gemini-2.5-pro

Verify ↗